Attention Projection Mixing with Exogenous Anchors

Jonathan Su

Keywords: Transformers, NLP, LLM

TL;DR: This paper proposes ExoFormer, a Transformer architecture that resolves the "first-layer tension" by learning dedicated exogenous anchor projections outside the main layer stack, which decouples token identity preservation from feature refinement.

Abstract: Cross-layer reuse of early attention projections can improve optimization and data efficiency, but it creates a structural conflict: the first layer must simultaneously act as a stable, reusable anchor for all deeper layers and as an effective computational block. We demonstrate that this tension constrains the performance of internal-anchor designs. We propose ExoFormer, which resolves the conflict by learning exogenous anchor projections outside the sequential layer stack. We introduce a unified normalized mixing framework that mixes queries, keys, values, and gate logits using learnable coefficients (exploring coefficient granularities: elementwise, headwise, and scalar), and we show that normalizing anchor sources is key to stable reuse. ExoFormer variants consistently outperform their internal-anchor counterparts, and the dynamic variant yields $\sim$1.5 downstream accuracy points while matching validation loss using $\sim$1.5x fewer tokens than Gated Attention. We explain this efficacy via an Offloading Hypothesis: external anchors preserve essential token identity, allowing layers to specialize exclusively in feature transformation. We release code and models to facilitate future research.

Anonymization: This submission has been anonymized for double-blind review via the removal of identifying information such as names, affiliations, and identifying URLs.

Style Files: I have used the style files.

Challenge: This submission is an entry to the science of DL improvement challenge.

Submission Number: 55